Market Basket Analysis

Introduction

Danielle has asked the team to perform a market basket analysis to help Blackwell’s board of directors better understand the clientele that Electronidex is currently serving and if Electronidex would be an optimal acquisition.

A dataset of transactions has been provided. The dataset contains 9835 transactions and 125 different products over a 30-day period, or about 327 transactions a day. This tells us the retailer is neither large, nor small.

Results and conclussions

After the analysis, we conclude Electronidex’s sales are to be categorized in two forms: retail (B2C) and corporate (B2B). Had we had this information previous to our analysis, it would have saved time in the exploration phase.

Interesting patterns and item relationships found: Retail: >

Would Blackwell benefit in selling Electronidex’s items?

Limitations and observations:

Properties of the dataset: 1. The iMac is the product most bought, in 20% of all transactions. This high number stands out considering the large variety of products, especially being the iMac a pricey product. If this number is representative of all sales throughout the year, then Electronidex is potentially profitable. 2. The mean of items bought per transaction is almost 5. Logically, I would say most people in the real world would buy 1 or 2 items per transactions most frequently, in an eletronics store.

Recommendations

Blackwell should acquire Electronidex If Blackwell does acquire Electronidex, do you have any recommendations for Blackwell? (Ex: cross-selling #items, sale promotions, should they remove items, etc.)

Rules for products in corporate transactions

rules_corpro <- apriori(trans_corp, parameter = list(supp = 0.01, conf = 0.01, 
    minlen = 2))

Apriori

Parameter specification:
 confidence minval smax arem  aval originalSupport maxtime support minlen
       0.01    0.1    1 none FALSE            TRUE       5    0.01      2
 maxlen target   ext
     10  rules FALSE

Algorithmic control:
 filter tree heap memopt load sort verbose
    0.1 TRUE TRUE  FALSE TRUE    2    TRUE

Absolute minimum support count: 58 

set item appearances ...[0 item(s)] done [0.00s].
set transactions ...[125 item(s), 5835 transaction(s)] done [0.00s].
sorting and recoding items ... [96 item(s)] done [0.00s].
creating transaction tree ... done [0.00s].
checking subsets of size 1 2 3 4 5 done [0.00s].
writing ... [1752 rule(s)] done [0.00s].
creating S4 object  ... done [0.00s].

rules_corpro[!is.redundant(rules_corpro)]

set of 1607 rules 

summary(rules_corpro)

set of 1752 rules

rule length distribution (lhs + rhs):sizes
   2    3    4 
1032  672   48 

   Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
  2.000   2.000   2.000   2.438   3.000   4.000 

summary of quality measures:
    support          confidence           lift            count      
 Min.   :0.01011   Min.   :0.02775   Min.   :0.6424   Min.   : 59.0  
 1st Qu.:0.01148   1st Qu.:0.12435   1st Qu.:1.0782   1st Qu.: 67.0  
 Median :0.01422   Median :0.20365   Median :1.2491   Median : 83.0  
 Mean   :0.01827   Mean   :0.23648   Mean   :1.3030   Mean   :106.6  
 3rd Qu.:0.01971   3rd Qu.:0.32164   3rd Qu.:1.4580   3rd Qu.:115.0  
 Max.   :0.12734   Max.   :0.69388   Max.   :2.7337   Max.   :743.0  

mining info:
       data ntransactions support confidence
 trans_corp          5835    0.01       0.01

# The most populars products
corpro_sup <- inspect(head(sort(rules_corpro, by = "support"), 10))

     lhs                           rhs                        support   
[1]  {HP Laptop}                => {iMac}                     0.12733505
[2]  {iMac}                     => {HP Laptop}                0.12733505
[3]  {Lenovo Desktop Computer}  => {iMac}                     0.09905741
[4]  {iMac}                     => {Lenovo Desktop Computer}  0.09905741
[5]  {CYBERPOWER Gamer Desktop} => {iMac}                     0.09562982
[6]  {iMac}                     => {CYBERPOWER Gamer Desktop} 0.09562982
[7]  {Dell Desktop}             => {iMac}                     0.09203085
[8]  {iMac}                     => {Dell Desktop}             0.09203085
[9]  {ViewSonic Monitor}        => {iMac}                     0.08157669
[10] {iMac}                     => {ViewSonic Monitor}        0.08157669
     confidence lift     count
[1]  0.4267662  1.132930 743  
[2]  0.3380346  1.132930 743  
[3]  0.4355690  1.156299 578  
[4]  0.2629663  1.156299 578  
[5]  0.3798502  1.008383 558  
[6]  0.2538672  1.008383 558  
[7]  0.4372964  1.160885 537  
[8]  0.2443130  1.160885 537  
[9]  0.4803229  1.275107 476  
[10] 0.2165605  1.275107 476  

# Items that have high chances of being bought together
corpro_con <- inspect(head(sort(rules_corpro, by = "confidence"), 10))

     lhs                          rhs            support confidence     lift count
[1]  {Dell Desktop,                                                               
      Lenovo Desktop Computer,                                                    
      ViewSonic Monitor}       => {iMac}      0.01165381  0.6938776 1.842027    68
[2]  {Apple Magic Keyboard,                                                       
      ASUS Monitor}            => {iMac}      0.01148243  0.6767677 1.796606    67
[3]  {Acer Aspire,                                                                
      iMac,                                                                       
      ViewSonic Monitor}       => {HP Laptop} 0.01045416  0.6630435 2.222205    61
[4]  {Acer Desktop,                                                               
      HP Laptop,                                                                  
      ViewSonic Monitor}       => {iMac}      0.01079692  0.6562500 1.742138    63
[5]  {ASUS 2 Monitor,                                                             
      Dell Desktop}            => {iMac}      0.01525278  0.6449275 1.712080    89
[6]  {ASUS Monitor,                                                               
      Lenovo Desktop Computer} => {iMac}      0.01645244  0.6442953 1.710402    96
[7]  {Acer Desktop,                                                               
      ASUS 2 Monitor}          => {iMac}      0.01079692  0.6428571 1.706584    63
[8]  {ASUS Monitor,                                                               
      ViewSonic Monitor}       => {iMac}      0.01388175  0.6428571 1.706584    81
[9]  {ASUS Monitor,                                                               
      Dell Desktop}            => {iMac}      0.01336761  0.6393443 1.697258    78
[10] {Acer Desktop,                                                               
      iMac,                                                                       
      ViewSonic Monitor}       => {HP Laptop} 0.01079692  0.6363636 2.132787    63

# Lift
corpro_lift <- inspect(head(sort(rules_corpro, by = "lift"), 10))

     lhs                              rhs                       
[1]  {HP Black & Tri-color Ink}    => {ViewSonic Monitor}       
[2]  {ViewSonic Monitor}           => {HP Black & Tri-color Ink}
[3]  {Acer Aspire,HP Laptop,iMac}  => {ViewSonic Monitor}       
[4]  {Dell 2 Desktop}              => {Apple Magic Keyboard}    
[5]  {Apple Magic Keyboard}        => {Dell 2 Desktop}          
[6]  {Apple Magic Keyboard,iMac}   => {ASUS Monitor}            
[7]  {Brother Printer}             => {iPad Pro}                
[8]  {iPad Pro}                    => {Brother Printer}         
[9]  {HP Laptop,ViewSonic Monitor} => {Computer Game}           
[10] {HP Wireless Mouse}           => {Epson Printer}           
     support    confidence lift     count
[1]  0.01113967 0.46428571 2.733711 65   
[2]  0.01113967 0.06559031 2.733711 65   
[3]  0.01045416 0.46212121 2.720966 61   
[4]  0.01439589 0.28378378 2.679415 84   
[5]  0.01439589 0.13592233 2.679415 84   
[6]  0.01148243 0.21824104 2.551977 67   
[7]  0.01028278 0.20547945 2.529478 60   
[8]  0.01028278 0.12658228 2.529478 60   
[9]  0.01233933 0.15652174 2.509078 72   
[10] 0.01165381 0.18630137 2.470610 68   

Rules for categories in corporate transactions

trans_corcat <- aggregate(trans_corp, by = "category")
rules_corcat <- apriori(trans_corcat, parameter = list(supp = 0.05, conf = 0.01, 
    minlen = 2))

Apriori

Parameter specification:
 confidence minval smax arem  aval originalSupport maxtime support minlen
       0.01    0.1    1 none FALSE            TRUE       5    0.05      2
 maxlen target   ext
     10  rules FALSE

Algorithmic control:
 filter tree heap memopt load sort verbose
    0.1 TRUE TRUE  FALSE TRUE    2    TRUE

Absolute minimum support count: 291 

set item appearances ...[0 item(s)] done [0.00s].
set transactions ...[17 item(s), 5835 transaction(s)] done [0.00s].
sorting and recoding items ... [17 item(s)] done [0.00s].
creating transaction tree ... done [0.00s].
checking subsets of size 1 2 3 4 5 done [0.00s].
writing ... [502 rule(s)] done [0.00s].
creating S4 object  ... done [0.00s].

rules_corcat[!is.redundant(rules_corcat)]

set of 479 rules 

summary(rules_corcat)

set of 502 rules

rule length distribution (lhs + rhs):sizes
  2   3   4   5 
136 228 128  10 

   Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
  2.000   2.000   3.000   3.024   4.000   5.000 

summary of quality measures:
    support          confidence          lift            count       
 Min.   :0.05004   Min.   :0.0763   Min.   :0.9674   Min.   : 292.0  
 1st Qu.:0.05947   1st Qu.:0.3226   1st Qu.:1.0542   1st Qu.: 347.0  
 Median :0.07232   Median :0.4639   Median :1.1448   Median : 422.0  
 Mean   :0.09491   Mean   :0.5123   Mean   :1.1463   Mean   : 553.8  
 3rd Qu.:0.10651   3rd Qu.:0.7248   3rd Qu.:1.2188   3rd Qu.: 621.5  
 Max.   :0.54070   Max.   :0.8735   Max.   :1.5310   Max.   :3155.0  

mining info:
         data ntransactions support confidence
 trans_corcat          5835    0.05       0.01

# The most populars categories
corcat_sup <- inspect(head(sort(rules_corcat, by = "support"), 10))

     lhs                   rhs        support   confidence lift     count
[1]  {Laptops}          => {Desktop}  0.5407027 0.7879620  1.008059 3155 
[2]  {Desktop}          => {Laptops}  0.5407027 0.6917343  1.008059 3155 
[3]  {Monitors}         => {Desktop}  0.4327335 0.7868495  1.006636 2525 
[4]  {Desktop}          => {Monitors} 0.4327335 0.5536067  1.006636 2525 
[5]  {Monitors}         => {Laptops}  0.3797772 0.6905578  1.006345 2216 
[6]  {Laptops}          => {Monitors} 0.3797772 0.5534466  1.006345 2216 
[7]  {Laptops,Monitors} => {Desktop}  0.3076264 0.8100181  1.036276 1795 
[8]  {Desktop,Monitors} => {Laptops}  0.3076264 0.7108911  1.035976 1795 
[9]  {Desktop,Laptops}  => {Monitors} 0.3076264 0.5689382  1.034514 1795 
[10] {Computer Mice}    => {Desktop}  0.2904884 0.7821874  1.000672 1695 

# Items that have high chances of being bought together
corcat_con <- inspect(head(sort(rules_corcat, by = "confidence"), 10))

     lhs                   rhs          support confidence     lift count
[1]  {Computer Mice,                                                     
      Keyboard,                                                          
      Laptops,                                                           
      Monitors}         => {Desktop} 0.06272494  0.8735084 1.117501   366
[2]  {Accessories,                                                       
      Keyboard,                                                          
      Monitors}         => {Desktop} 0.05895458  0.8600000 1.100219   344
[3]  {Keyboard,                                                          
      Laptops,                                                           
      Monitors}         => {Desktop} 0.12613539  0.8578089 1.097416   736
[4]  {Accessories,                                                       
      Keyboard,                                                          
      Laptops}          => {Desktop} 0.06152528  0.8568019 1.096128   359
[5]  {Computer Mice,                                                     
      Computer Tablets} => {Desktop} 0.05604113  0.8560209 1.095129   327
[6]  {Computer Mice,                                                     
      Keyboard,                                                          
      Monitors}         => {Desktop} 0.07866324  0.8531599 1.091468   459
[7]  {Accessories,                                                       
      Laptops,                                                           
      Monitors}         => {Desktop} 0.09340189  0.8475894 1.084342   545
[8]  {Computer Cords,                                                    
      Laptops,                                                           
      Monitors}         => {Desktop} 0.05621251  0.8475452 1.084286   328
[9]  {Accessories,                                                       
      Computer Mice,                                                     
      Monitors}         => {Desktop} 0.05724079  0.8455696 1.081758   334
[10] {Computer Mice,                                                     
      Keyboard,                                                          
      Laptops}          => {Desktop} 0.09048843  0.8448000 1.080774   528

# Lift
corcat_lift <- inspect(head(sort(rules_corcat, by = "lift"), 10))

     lhs                                 rhs                       
[1]  {Keyboard,Laptops,Monitors}      => {Accessories}             
[2]  {Desktop,Keyboard,Monitors}      => {Accessories}             
[3]  {Keyboard,Monitors}              => {Accessories}             
[4]  {Computer Mice,Laptops,Monitors} => {Accessories}             
[5]  {Computer Cords,Desktop}         => {Keyboard}                
[6]  {Desktop,Keyboard,Laptops}       => {Accessories}             
[7]  {Computer Mice,Desktop,Monitors} => {Accessories}             
[8]  {Accessories,Desktop,Monitors}   => {Keyboard}                
[9]  {Computer Cords}                 => {Mouse and Keyboard Combo}
[10] {Mouse and Keyboard Combo}       => {Computer Cords}          
     support    confidence lift     count
[1]  0.05158526 0.3508159  1.531047 301  
[2]  0.05895458 0.3412698  1.489386 344  
[3]  0.06855184 0.3276003  1.429729 400  
[4]  0.05347044 0.3260188  1.422827 312  
[5]  0.05141388 0.4807692  1.398449 300  
[6]  0.06152528 0.3182624  1.388976 359  
[7]  0.05724079 0.3177926  1.386926 334  
[8]  0.05895458 0.4744828  1.380163 344  
[9]  0.05107112 0.3692689  1.364588 298  
[10] 0.05107112 0.1887270  1.364588 298  

Rules for products in retailers transactions

rules_retpro <- apriori(trans_retail, parameter = list(supp = 0.005, conf = 0.01, 
    minlen = 2))

Apriori

Parameter specification:
 confidence minval smax arem  aval originalSupport maxtime support minlen
       0.01    0.1    1 none FALSE            TRUE       5   0.005      2
 maxlen target   ext
     10  rules FALSE

Algorithmic control:
 filter tree heap memopt load sort verbose
    0.1 TRUE TRUE  FALSE TRUE    2    TRUE

Absolute minimum support count: 20 

set item appearances ...[0 item(s)] done [0.00s].
set transactions ...[125 item(s), 4000 transaction(s)] done [0.00s].
sorting and recoding items ... [68 item(s)] done [0.00s].
creating transaction tree ... done [0.00s].
checking subsets of size 1 2 done [0.00s].
writing ... [22 rule(s)] done [0.00s].
creating S4 object  ... done [0.00s].

rules_retpro[!is.redundant(rules_retpro)]

set of 22 rules 

summary(rules_retpro)

set of 22 rules

rule length distribution (lhs + rhs):sizes
 2 
22 

   Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
      2       2       2       2       2       2 

summary of quality measures:
    support           confidence           lift            count      
 Min.   :0.005000   Min.   :0.04600   Min.   :0.5283   Min.   :20.00  
 1st Qu.:0.006375   1st Qu.:0.06696   1st Qu.:1.3676   1st Qu.:25.50  
 Median :0.007250   Median :0.10666   Median :1.4819   Median :29.00  
 Mean   :0.007932   Mean   :0.13853   Mean   :1.8501   Mean   :31.73  
 3rd Qu.:0.009062   3rd Qu.:0.19223   3rd Qu.:2.4736   3rd Qu.:36.25  
 Max.   :0.014250   Max.   :0.45763   Max.   :3.6096   Max.   :57.00  

mining info:
         data ntransactions support confidence
 trans_retail          4000   0.005       0.01

# The most populars products
retpro_sup <- inspect(head(sort(rules_retpro, by = "support"), 10))

     lhs                           rhs                        support
[1]  {CYBERPOWER Gamer Desktop} => {Apple Earpods}            0.01425
[2]  {Apple Earpods}            => {CYBERPOWER Gamer Desktop} 0.01425
[3]  {Apple MacBook Air}        => {Apple Earpods}            0.00975
[4]  {Apple Earpods}            => {Apple MacBook Air}        0.00975
[5]  {3-Button Mouse}           => {Apple Earpods}            0.00925
[6]  {Apple Earpods}            => {3-Button Mouse}           0.00925
[7]  {Samsung Monitor}          => {CYBERPOWER Gamer Desktop} 0.00850
[8]  {CYBERPOWER Gamer Desktop} => {Samsung Monitor}          0.00850
[9]  {Apple Macbook Pro}        => {Apple Earpods}            0.00800
[10] {Apple Earpods}            => {Apple Macbook Pro}        0.00800
     confidence lift      count
[1]  0.16764706 1.3331774 57   
[2]  0.11332008 1.3331774 57   
[3]  0.06643952 0.5283461 39   
[4]  0.07753479 0.5283461 39   
[5]  0.20000000 1.5904573 37   
[6]  0.07355865 1.5904573 37   
[7]  0.23129252 2.7210884 34   
[8]  0.10000000 2.7210884 34   
[9]  0.21768707 1.7311099 32   
[10] 0.06361829 1.7311099 32   

# Items that have high chances of being bought together
retpro_con <- inspect(head(sort(rules_retpro, by = "confidence"), 10))

     lhs                                                rhs                        support confidence     lift count
[1]  {iPhone Charger Cable}                          => {Apple MacBook Air}        0.00675  0.4576271 3.118413    27
[2]  {Acer Monitor}                                  => {CYBERPOWER Gamer Desktop} 0.00675  0.3068182 3.609626    27
[3]  {Samsung Monitor}                               => {CYBERPOWER Gamer Desktop} 0.00850  0.2312925 2.721088    34
[4]  {Apple Macbook Pro}                             => {Apple Earpods}            0.00800  0.2176871 1.731110    32
[5]  {Microsoft Wireless Desktop Keyboard and Mouse} => {Apple MacBook Air}        0.00725  0.2132353 1.453051    29
[6]  {3-Button Mouse}                                => {Apple Earpods}            0.00925  0.2000000 1.590457    37
[7]  {Dell Laptop}                                   => {Apple Earpods}            0.00625  0.1689189 1.343292    25
[8]  {CYBERPOWER Gamer Desktop}                      => {Apple Earpods}            0.01425  0.1676471 1.333177    57
[9]  {3-Button Mouse}                                => {iMac}                     0.00550  0.1189189 1.481856    22
[10] {Backlit LED Gaming Keyboard}                   => {iMac}                     0.00500  0.1156069 1.440585    20

# Lift
retpro_lift <- inspect(head(sort(rules_retpro, by = "lift"), 10))

     lhs                           rhs                        support
[1]  {Acer Monitor}             => {CYBERPOWER Gamer Desktop} 0.00675
[2]  {CYBERPOWER Gamer Desktop} => {Acer Monitor}             0.00675
[3]  {iPhone Charger Cable}     => {Apple MacBook Air}        0.00675
[4]  {Apple MacBook Air}        => {iPhone Charger Cable}     0.00675
[5]  {Samsung Monitor}          => {CYBERPOWER Gamer Desktop} 0.00850
[6]  {CYBERPOWER Gamer Desktop} => {Samsung Monitor}          0.00850
[7]  {Apple Macbook Pro}        => {Apple Earpods}            0.00800
[8]  {Apple Earpods}            => {Apple Macbook Pro}        0.00800
[9]  {3-Button Mouse}           => {Apple Earpods}            0.00925
[10] {Apple Earpods}            => {3-Button Mouse}           0.00925
     confidence lift     count
[1]  0.30681818 3.609626 27   
[2]  0.07941176 3.609626 27   
[3]  0.45762712 3.118413 27   
[4]  0.04599659 3.118413 27   
[5]  0.23129252 2.721088 34   
[6]  0.10000000 2.721088 34   
[7]  0.21768707 1.731110 32   
[8]  0.06361829 1.731110 32   
[9]  0.20000000 1.590457 37   
[10] 0.07355865 1.590457 37   

Rules for categories in retailers transactions

trans_retcat <- aggregate(trans_retail, by = "category")
rules_retcat <- apriori(trans_retcat, parameter = list(supp = 0.01, conf = 0.01, 
    minlen = 2))

Apriori

Parameter specification:
 confidence minval smax arem  aval originalSupport maxtime support minlen
       0.01    0.1    1 none FALSE            TRUE       5    0.01      2
 maxlen target   ext
     10  rules FALSE

Algorithmic control:
 filter tree heap memopt load sort verbose
    0.1 TRUE TRUE  FALSE TRUE    2    TRUE

Absolute minimum support count: 40 

set item appearances ...[0 item(s)] done [0.00s].
set transactions ...[17 item(s), 4000 transaction(s)] done [0.00s].
sorting and recoding items ... [17 item(s)] done [0.00s].
creating transaction tree ... done [0.00s].
checking subsets of size 1 2 3 done [0.00s].
writing ... [40 rule(s)] done [0.00s].
creating S4 object  ... done [0.00s].

rules_retpro[!is.redundant(rules_retpro)]

set of 22 rules 

summary(rules_retcat)

set of 40 rules

rule length distribution (lhs + rhs):sizes
 2 
40 

   Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
      2       2       2       2       2       2 

summary of quality measures:
    support          confidence           lift            count    
 Min.   :0.01025   Min.   :0.03275   Min.   :0.5645   Min.   : 41  
 1st Qu.:0.01250   1st Qu.:0.08498   1st Qu.:0.6138   1st Qu.: 50  
 Median :0.01675   Median :0.12546   Median :0.7535   Median : 67  
 Mean   :0.02275   Mean   :0.15262   Mean   :0.8038   Mean   : 91  
 3rd Qu.:0.03625   3rd Qu.:0.21769   3rd Qu.:0.9706   3rd Qu.:145  
 Max.   :0.05375   Max.   :0.35602   Max.   :1.2459   Max.   :215  

mining info:
         data ntransactions support confidence
 trans_retcat          4000    0.01       0.01

# The most populars products
retcat_sup <- inspect(head(sort(rules_retcat, by = "support"), 10))

     lhs                    rhs                 support confidence
[1]  {Monitors}          => {Desktop}           0.05375 0.3272451 
[2]  {Desktop}           => {Monitors}          0.05375 0.1881015 
[3]  {Computer Mice}     => {Desktop}           0.03825 0.2875940 
[4]  {Desktop}           => {Computer Mice}     0.03825 0.1338583 
[5]  {Monitors}          => {Laptops}           0.03725 0.2267884 
[6]  {Laptops}           => {Monitors}          0.03725 0.1190096 
[7]  {Keyboard}          => {Desktop}           0.03625 0.2859961 
[8]  {Desktop}           => {Keyboard}          0.03625 0.1268591 
[9]  {Active Headphones} => {Laptops}           0.03625 0.2449324 
[10] {Laptops}           => {Active Headphones} 0.03625 0.1158147 
     lift      count
[1]  1.1452145 215  
[2]  1.1452145 215  
[3]  1.0064531 153  
[4]  1.0064531 153  
[5]  0.7245637 149  
[6]  0.7245637 149  
[7]  1.0008611 145  
[8]  1.0008611 145  
[9]  0.7825317 145  
[10] 0.7825317 145  

# Items that have high chances of being bought together
retcat_con <- inspect(head(sort(rules_retcat, by = "confidence"), 10))

     lhs                           rhs       support confidence lift     
[1]  {Accessories}              => {Desktop} 0.01700 0.3560209  1.2459176
[2]  {Monitors}                 => {Desktop} 0.05375 0.3272451  1.1452145
[3]  {Mouse and Keyboard Combo} => {Laptops} 0.02275 0.3204225  1.0237142
[4]  {Computer Cords}           => {Laptops} 0.01150 0.3006536  0.9605546
[5]  {Computer Mice}            => {Desktop} 0.03825 0.2875940  1.0064531
[6]  {Keyboard}                 => {Desktop} 0.03625 0.2859961  1.0008611
[7]  {Computer Headphones}      => {Desktop} 0.01250 0.2577320  0.9019491
[8]  {Active Headphones}        => {Laptops} 0.03625 0.2449324  0.7825317
[9]  {Active Headphones}        => {Desktop} 0.03625 0.2449324  0.8571564
[10] {Monitors}                 => {Laptops} 0.03725 0.2267884  0.7245637
     count
[1]   68  
[2]  215  
[3]   91  
[4]   46  
[5]  153  
[6]  145  
[7]   50  
[8]  145  
[9]  145  
[10] 149  

# Lift
retcat_lift <- inspect(head(sort(rules_retcat, by = "lift"), 10))

     lhs                           rhs                        support
[1]  {Accessories}              => {Desktop}                  0.01700
[2]  {Desktop}                  => {Accessories}              0.01700
[3]  {Desktop}                  => {Monitors}                 0.05375
[4]  {Monitors}                 => {Desktop}                  0.05375
[5]  {Mouse and Keyboard Combo} => {Laptops}                  0.02275
[6]  {Laptops}                  => {Mouse and Keyboard Combo} 0.02275
[7]  {Computer Mice}            => {Desktop}                  0.03825
[8]  {Desktop}                  => {Computer Mice}            0.03825
[9]  {Keyboard}                 => {Desktop}                  0.03625
[10] {Desktop}                  => {Keyboard}                 0.03625
     confidence lift     count
[1]  0.35602094 1.245918  68  
[2]  0.05949256 1.245918  68  
[3]  0.18810149 1.145215 215  
[4]  0.32724505 1.145215 215  
[5]  0.32042254 1.023714  91  
[6]  0.07268371 1.023714  91  
[7]  0.28759398 1.006453 153  
[8]  0.13385827 1.006453 153  
[9]  0.28599606 1.000861 145  
[10] 0.12685914 1.000861 145